CN111307812A - Appearance inspection method of solder joints based on machine vision - Google Patents

Abstract

The invention discloses a method for inspecting the appearance of solder joints based on machine vision, belonging to the technical field of appearance inspection of solder joints. The method for inspecting appearance of solder joints based on machine vision comprises the following steps: firstly, a workpiece to be inspected is fixed on a conveyor belt, and technicians adjust the workpiece And the position of the camera, make the high-definition camera close to the workpiece welding point, and then the high-definition camera exposes the workpiece for multiple times to obtain multiple high-definition images of the welding point and transmit it to the computer, and then process the images, and the image analysis module will process the The image of the workpiece and the template image are compared and analyzed. When the difference between the workpiece image and the template is large, the image analysis module sends the alarm module, and the image analysis module sends the analysis results to the technicians, which can facilitate the welding of the workpiece image. Accurate analysis results of the appearance of the solder joints can be obtained by comparing and analyzing the solder joint images of different sizes with the corresponding templates.

Description

Appearance inspection method of solder joints based on machine vision

technical field

The invention relates to the technical field of appearance inspection of solder joints, and more particularly, to a method for appearance inspection of solder joints based on machine vision.

Background technique

With the rapid development of my country's manufacturing, construction, water conservancy, environment and other industries and the continuous growth of foreign trade, my country's quality inspection and testing industry has also ushered in rapid development. Machine vision system refers to converting the captured target into image signals through machine vision products, such as CCD, CMOS and photocells, and sending them to a dedicated image processing system. Signal; the image system performs various operations on these signals to extract the characteristics of the target, and then controls the equipment on site according to the result of the discrimination. It can improve product quality, reduce scrap rate, and improve enterprise economic benefits. The improvement of the living standards of Chinese residents will continue to drive the demand for consumer goods and life science testing to rise, and the rapid development of the manufacturing industry and industrial upgrading will boost the demand for industrial product testing. However, the traditional manual visual inspection method has low inspection efficiency, and the inspection results and quality are related to the experience of the inspectors, so the inspection results are inaccurate.

Spot welding machines are mainly used in equipment in the electronics industry and have high quality requirements. Therefore, the quality of solder joints needs to be tested. Due to the immaturity of the actual process and the mistakes of manual operation, there are often some defects, such as too much solder, too little solder, rosin soldering, overheating soldering, cold soldering, virtual soldering, asymmetry, looseness, sharpening, bridging, pinhole etc. Common defects are mainly virtual welding and welding through. The reasons for the virtual welding are generally: the pressure of the electrodes on both sides is too small, so that the weldment is not well attached, and the current is too small. The main reason for the weld penetration is that the current is too large, which directly leads to the perforation of the weldment.

Most of the solder joint defect detection and classification algorithms proposed in the prior art have low detection efficiency and high misjudgment rate; or algorithm parameters need to be set manually and have poor stability, and the calculation complexity is relatively large.

SUMMARY OF THE INVENTION

1. Technical problems to be solved

In view of the problems existing in the prior art, the purpose of the present invention is to provide a method for detecting the appearance of solder joints based on machine vision, which can realize the acquisition of solder joint images by using a solder joint image acquisition device, which facilitates the collection of solder joint images of the workpiece and facilitates the acquisition of solder joint images. For solder joint images of different sizes, by comparing and analyzing the solder joint images of different sizes with multiple templates, accurate solder joint appearance analysis results can be obtained.

2. Technical solutions

In order to solve the above problems, the present invention adopts the following technical solutions.

A solder joint appearance inspection method based on machine vision includes an appearance inspection system, the appearance inspection system includes a control module, and the control module is connected with an image acquisition module, a light source control module, a camera position adjustment module, an image processing module and an image analysis module. The image analysis module is connected with an alarm module, and the control module, the image processing module, the image analysis module and the alarm module are all installed in the computer.

Further, the image acquisition module is a high-definition camera, a wireless signal transmitter is connected to the upper end of the high-definition camera, an LED lighting strip is connected to the high-definition camera, and the camera position adjustment module includes a linear guide rail and a lift, so The high-definition camera is connected with the linear guide rail, the lower end of the elevator is connected with a base, the base is connected with a conveyor belt matching the high-definition camera, and a transmission shaft of the conveyor belt is connected with a motor.

Further, its detection method is, comprising the following steps:

A1. Adjust the position of the workpiece and the image acquisition module, so that the camera is close to the welding point of the workpiece;

A2. Use multiple exposures through the image acquisition module to obtain high-definition images at the solder joints;

A3. The wireless signal transmitter transmits the obtained high-definition image to the computer, and the image processing module in the computer processes the image;

A4. Compare and analyze the image processed in the previous step and the template image through the image analysis module;

A5. Send the analysis results, and the technicians will evaluate the analysis results.

Further, the preparation of the template image in the A4 includes the following steps:

B1. Adjust the position of the standard workpiece and the image acquisition module, so that the camera is close to the welding point of the standard workpiece;

B2. Use multiple exposures through the image acquisition module to obtain high-definition images at multiple solder joints;

The B3 wireless signal transmitter transmits the obtained high-definition image to the computer, and the image processing module in the computer processes the image;

B4. Store the processed image in the image analysis module.

Further, the image processing work in A3 and B3 includes the following steps:

B201. Fusion of images obtained by multiple exposures;

B202. Adjust the contrast, saturation and brightness of the image to make the image clearer;

B203. Image denoising, remove noise in images;

B204. Image filtering; further removal of noise in images

B205. Image segmentation, select areas with abnormal grayscale.

Further, the light source control module includes a photosensitive sensor and an LED light strip.

Further, a light absorbing layer is laid on the conveyor belt, so that the conveyor belt is not easy to reflect light.

Further, a storage unit is connected in the control module, and the storage unit is connected with a cloud server.

Further, the solder joint image acquisition device is placed in a soundproof dark box, which is easy to reduce the influence of ambient light and ambient noise on the detection result.

3. Beneficial effects

Compared with the prior art, the advantages of the present invention are:

(1) The present invention uses a solder joint image acquisition device to collect solder joint images, which facilitates the collection of solder joint images of the workpiece and facilitates the acquisition of solder joint images of different sizes. By comparing and analyzing the solder joint images of different sizes with multiple templates, Accurate solder joint appearance analysis results can be obtained.

(2) The present invention controls the conveyor belt, the elevator and the linear guide rail to adjust the position of the workpiece and the camera according to the welding point positioning technology, so that the camera is directly facing the welding point, so as to obtain an accurate and clear image of the welding point. The welding point image acquisition device of the present invention uses Wireless signal control can realize that one computer controls multiple solder joint image acquisition devices to work, realizes multiple solder joint image acquisition devices to work at the same time, inspects multiple workpieces, and improves work efficiency.

(3) The present invention can upload the analysis results to the cloud server, so that the technicians can call at any time to analyze and evaluate the workpieces, and the workpieces with solder joint defects will trigger an alarm, so that the technicians can find the defective workpieces in time.

Description of drawings

Fig. 1 is the principle diagram of the appearance detection system of the present invention;

Fig. 2 is the flow chart of the solder joint detection method of the present invention;

Fig. 3 is the image processing step of the present invention;

4 is a perspective view of a solder joint image acquisition device of the present invention;

5 is a side view of the solder joint image acquisition device of the present invention;

Fig. 6 is the top view of the solder joint image acquisition device of the present invention;

FIG. 7 is a perspective view of the solder joint image acquisition device of the present invention when it is located in a dark box and works.

1 control module, 2 image acquisition module, 3 light source control, 4 camera position adjustment module, 401 linear guide rail, 402 elevator, 5 image processing module, 6 image analysis module, 7 alarm module, 8 conveyor belt, 9 wireless signal transmitter.

Detailed ways

The following will combine the drawings in the embodiments of the present invention; the technical solutions in the embodiments of the present invention will be described clearly and completely; obviously; the described embodiments are only a part of the embodiments of the present invention; rather than all the embodiments, based on The embodiments of the present invention; all other embodiments obtained by those of ordinary skill in the art without creative work; all belong to the protection scope of the present invention.

In the description of the present invention, it should be noted that the orientations or positional relationships indicated by the terms "upper", "lower", "inner", "outer", "top/bottom", etc. are based on the orientations shown in the accompanying drawings Or the positional relationship is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed to indicate or imply relative importance.

In the description of the present invention, it should be noted that, unless otherwise expressly specified and limited, the terms "installation", "provided with", "sleeve/connection", "connection", etc., should be understood in a broad sense, such as " Connection", which can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be an internal connection between two components. of connectivity. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood in specific situations.

Example 1:

Referring to FIG. 1, a method for inspecting the appearance of solder joints based on machine vision includes an appearance inspection system. The appearance inspection system includes a control module 1. The control module 1 is connected with an image acquisition module 2, a light source control module 3, and a camera position adjustment module 4. The image processing module 5 and the image analysis module 6, the image analysis module 6 is connected with the alarm module 7, the control module 1, the image processing module 5, the image analysis module 6 and the alarm module 7 are all installed in the computer, and the control module 1 is connected with The storage unit is connected with the cloud server.

Please refer to Figure 4-7, the image acquisition module 2 is connected to the solder joint image acquisition device, the image acquisition module 2 is a high-definition camera, the upper end of the high-definition camera is connected with a wireless signal transmitter 9, and the high-definition camera is connected with a light source control module 3, The light source control module 3 includes a photosensitive sensor and an LED light strip. Four LED light strips surround the high-definition camera. The camera position adjustment module 4 is signally connected to the wireless signal transmitter 9. The camera position adjustment module 4 includes a linear guide 401 and a lift 402. The high-definition camera Fixed connection with the slider on the linear guide 401, the high-definition camera can move along the linear guide 401, the linear guide 401 is connected to the elevator 402, the elevator 402 can make the linear guide 401 move in the vertical direction, the lower end of the elevator 402 is connected with a base, the base A conveyor belt 8 that matches the high-definition camera is connected to the top. The lens of the high-definition camera is parallel to the surface of the conveyor belt 8. A transmission shaft of the conveyor belt 8 is connected with a motor. The motor can drive the transmission shaft to move, thereby driving the conveyor belt to move. There is a light-absorbing layer, so that the conveyor belt 8 is not easy to reflect light, and the light-absorbing layer is used to reduce light reflection, so as to improve the quality of pictures taken by high-definition cameras. When the environment is brighter, the light sensitive sensor sends a feedback signal to the LED light strip to reduce the brightness of the LED light strip.

The motor can drive the conveyor belt 8 to move, and the high-definition camera can move along the linear guide 401, so that the high-definition camera is close to the welding spot on the workpiece. The technician can control the linear guide 401 and the motor according to the welding spot positioning technology, so as to adjust the high-definition camera. and the position of the workpiece, so that the welding point on the workpiece is facing the lens on the high-definition camera, the technician can also control the lift 402 to adjust the height of the high-definition camera, so that the high-definition camera can be located at different heights to shoot the workpiece, so that the high-definition camera can be The height can be adapted to the size of the workpiece, and images of solder joints of different sizes can be taken at different heights. The technicians can also put the whole device into a soundproof obscure box to reduce the interference of ambient light and ambient noise.

Please refer to Fig. 2, the inspection method for the appearance of solder joints based on machine vision, the inspection method is as follows: firstly, the workpiece to be inspected is fixed on the conveyor belt 8, the technician adjusts the position of the workpiece by using the motor to control the conveyor belt 8, and then uses the camera to adjust the position Module 4 adjusts the position of the high-definition camera, so that the high-definition camera is close to the welding spot of the workpiece; then the high-definition camera exposes the workpiece for multiple times to obtain high-definition images at multiple welding spots; the wireless signal transmitter 9 transmits the obtained high-definition images to the computer, The image processing module 5 in the computer processes the image; then the image analysis module 6 compares and analyzes the processed image and the template image; when the difference between the workpiece image and the template is larger, the image analysis module 6 sends to the alarm module 7, The alarm module 7 is made to issue an alarm to remind the technicians, and the image analysis module 6 sends the analysis results to the technicians for further analysis and evaluation. Those skilled in the art can adjust the size and scope of the difference according to the actual situation. The analysis module 6 will also upload the analysis results to the storage module, and the storage module will upload the received analysis results to the cloud server for backup, which is convenient for technicians to retrieve.

The preparation steps of the template image in the detection method are: adjusting the position of the standard workpiece and the image acquisition module 2, so that the camera is close to the welding point of the standard workpiece; using the image acquisition module 2 for multiple exposures to obtain high-definition images at multiple welding points; wireless signal The transmitter 9 transmits the obtained high-definition image to the computer, and the image processing module 5 in the computer processes the image; the processed image is stored in the image analysis module 6 .

The solder joint images of the same workpiece standard part may be different. The technician can use this method to input the solder joint images of multiple standard workpieces with different pixel sizes and sizes into the image analysis module 6, so that the image analysis module 6 can detect the workpiece when it is working. The images of different pixels and the images of the corresponding pixels and sizes of multiple templates are compared and analyzed and averaged to reduce the analysis error. In the image analysis, the perceptual hash algorithm can also be used to compare the image similarity to analyze the solder joint images. The size and size of the image can be adjusted by adjusting the height of the camera and the magnification of the camera.

Referring to Figure 3, the steps of image processing in the above detection method are: first, fuse the images obtained by multiple exposures; then adjust the contrast, saturation and brightness of the image to make the image clearer; and then denoise the image and filtering to eliminate noise in digital images; finally, the image is divided into multiple regions by region growing method according to the gray level, and the region with abnormal gray level is selected.

Through multiple exposure fusion, the image can achieve the effect of HDR, improve the dynamic range, and make the image more close to the real light effect. It is often polluted by a variety of noises. In addition, in some aspects of image processing, when the input image object is not as expected, noise will also be introduced into the resulting image. These noises often appear as an isolated pixel point or pixel block that causes a strong visual effect on the image. Generally, the noise signal is irrelevant to the object to be studied. It appears in the form of useless information, disturbing the observable information of the image. For digital image signals, the noise table is a large or small extreme value. These extreme values act on the real gray value of the image pixels through addition and subtraction, causing bright and dark spot interference to the image, greatly reducing the image quality and affecting the Image restoration, segmentation, feature extraction, image recognition and other follow-up work, denoising and filtering the image, can reduce the noise in the digital image, remove the noise in the target and the background, and can well protect the image target. Shape, size, and specific geometric and topological features.

The invention facilitates the collection of the welding spot images of the workpiece, so as to obtain the welding spot images of different sizes. It will be uploaded to the cloud server, so that the technician can call it at any time to analyze and evaluate the workpiece. The workpiece with solder joint defects will trigger the alarm module 7, so that the technician can find the defective workpiece in time. The present invention is based on the solder joint positioning technology. Control the conveyor belt 8, the elevator 401 and the linear guide 402 to adjust the position of the workpiece and the camera, so that the camera is directly facing the solder joint to obtain an accurate and clear solder joint image. The solder joint image acquisition device of the present invention uses wireless signal control, which can realize a A computer controls multiple solder point image acquisition devices to work, realizes multiple solder point image acquisition devices to work at the same time, inspects multiple workpieces, and improves work efficiency.

The above description is only a preferred embodiment of the present invention; however, the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical scope of the present invention, according to the technical solution of the present invention and its improvement concept, equivalently replaces or changes, should be covered within the protection scope of the present invention.

Claims (9)

1. The welding spot appearance detection method based on machine vision comprises an appearance detection system and is characterized in that: the appearance detection system comprises a control module (1), wherein the control module (1) is connected with an image acquisition module (2), a light source control module (3), a camera position adjusting module (4), an image processing module (5) and an image analysis module (6), the image analysis module (6) is connected with an alarm module (7), and the control module (1), the image processing module (5), the image analysis module (6) and the alarm module (7) are all installed in a computer.

2. The machine vision-based weld spot appearance detection method according to claim 1 or 2, wherein: including solder joint image acquisition device, image acquisition module (2) are connected on solder joint image acquisition device, image acquisition module (2) are the high definition camera, the upper end of high definition camera is connected with wireless signal transmitter (9), be connected with light source control module (3) on the high definition camera, camera position control module (4) include linear guide (401) and lift (402), the high definition camera is connected with linear guide (401), the lower extreme of lift (402) is connected with the base, be connected with on the base with high definition camera assorted conveyer belt (8), be connected with the motor on a transmission shaft of conveyer belt (8).

3. The machine vision-based weld spot appearance detection method according to claim 1 or 2, wherein: the detection method comprises the following steps:

A1. adjusting the positions of the workpiece and the image acquisition module (2) to enable the camera to be close to a welding point of the workpiece;

A2. obtaining a high-definition image at a welding point by using multiple exposures through the image acquisition module (2);

A3. the wireless signal transmitter (9) transmits the obtained high-definition image to a computer, and an image processing module (5) in the computer processes the image;

A4. comparing and analyzing the image processed in the previous step and the template image through an image analysis module (6);

A5. and sending the analysis result, and evaluating the analysis result by a technician.

4. The machine vision-based weld spot appearance detection method according to claim 1 or 3, wherein: the preparation of the template image in a4 comprises the following steps:

B1. adjusting the positions of the standard workpiece and the image acquisition module (2) to enable the camera to be close to the welding point of the standard workpiece;

B2. obtaining high-definition images at a plurality of welding points by using multiple exposures of the image acquisition module (2);

b3 the wireless signal transmitter (9) transmits the obtained high-definition image to the computer, and the image processing module (5) in the computer processes the image;

B4. and storing the processed image into an image analysis module (6).

5. The machine vision-based weld spot appearance detection method according to claim 1 or 3, wherein: the image processing work in A3 and B3 includes the following steps:

B201. fusing images obtained by multiple exposures;

B202. adjusting the contrast, saturation and brightness of the image to make the image clearer;

B203. denoising the image, namely eliminating noise points in the image;

B204. filtering the image to further eliminate noise in the image;

B205. and (5) image segmentation, namely selecting an area with abnormal gray scale.

6. The machine vision-based weld spot appearance detection method according to claim 1, wherein: the light source control module (3) comprises a photosensitive sensor and an LED lamp strip.

7. The machine vision-based weld spot appearance detection method according to claim 1 or 2, wherein: and a light absorption layer is paved on the conveyor belt (8).

8. The machine vision-based weld spot appearance detection method according to claim 1, wherein: the cloud server is characterized in that a storage unit is connected in the control module (1) and is connected with the cloud server.

9. The machine vision-based weld spot appearance detection method according to claim 1, wherein: the welding spot image acquisition device is placed in a sound insulation dark box.

Images